Morning Brief

Routine cybersecurity testing of frontier AI models sparked a series of unexpected security incidents—the most serious case arising when Anthropic’s Mythos 5 model attempted to insert malicious code into an open source software application and created fake identities to deceive the human developers maintaining the project. The security incidents occurred during a cyber evaluation of seven leading AI models’ capabilities by the AI Security Institute (AISI), a research organization within the UK government, in late July. The researchers discovered 19 instances in which “AI agents took…

Cybercriminals spent July 2026 proving that trusted business utilities including Microsoft authentication pages, Zoom event invitations, and official government portals can be weaponized against enterprise targets. Threat intelligence research from ANY.RUN reveals that attackers across the United States, Europe, and Brazil systematically exploited routine corporate workflows to bypass perimeter security controls, harvest credentials, and maintain long-term access to systems. Hackers Turn Trusted Sites Into Attack Tools The defining trend across July’s threat landscape was the exploitation of…

A newly patched security flaw impacting on-premise versions of JetBrains TeamCity has come under active exploitation in the wild, according to the U.S. Cybersecurity and Infrastructure Security Agency (CISA). The vulnerability in question is CVE-2026-63077 (CVSS score: 9.8), a case of deserialization of untrusted data that could allow an unauthenticated attacker with access to a TeamCity server

Apple users who rely on iCloud Private Relay to conceal their online location may not be getting the protection they expect. Newly disclosed flaws in WebKit, the browser engine used across iOS, can allow a website to see a visitor’s real IP address. The exposure weakens a privacy safeguard. The issue can be triggered when a site supports, or merely claims to support, passkeys. A device may make a separate network request during sign-in, outside the protected browser route. That creates an opening for a malicious site to collect the address. Researchers Tommy Mysk and Talal Haj Bakry…

A critical vulnerability in IBM-owned, low-code AI builder Langflow lets unauthenticated attackers execute code remotely on vulnerable default deployments, potentially putting organizations running those instances at immediate risk. The Cybersecurity and Infrastructure Security Agency (CISA) on Tuesday added CVE-2026-9198 to its Known Exploited Vulnerabilities catalog after identifying evidence of active exploitation and urged organizations to apply the vendor's mitigation guidance as soon as possible. IBM says the flaw affects Langflow OSS versions 1.0.0 through 1.10.0 and recommends…

Atomic Stealer is being pushed at Mac users through websites that look harmless. A large ClickFix operation uses more than 250 look-alike domains to steer selected visitors toward a fake download page and a malicious Terminal command. ClickFix is built on persuasion, not a software flaw. The pages pose as download checks, updates, or CAPTCHA-style verifications, then urge people to paste a command into Terminal to begin the attack. Microsoft analysts identified this shift. They found that the infrastructure moved from openly displaying malicious instructions to testing each visitor before…

A federal judge in Alexandria, Virginia, sentenced Maksim Silnikau to 16 years in prison on August 5 for creating and running Ransom Cartel, the ransomware-as-a-service operation he stood up in 2021. Between 2021 and 2023, Ransom Cartel conspirators attacked at least 18 companies, including firms in California, New York and Nebraska, and others abroad, according to the Justice Department.

Cisco has released a critical security hardening update for Cisco IOS XE Software, fixing several serious vulnerabilities that could expose enterprise network devices to remote attacks. The advisory covers vulnerabilities identified during Cisco’s internal security testing, including testing supported by frontier AI models. Cisco said it is not aware of public exploitation or malicious activity linked to these flaws. However, the severity of the issues, combined with the lack of available workarounds, makes prompt patching essential. The vulnerabilities affect Cisco IOS XE Software running in…

Security flaws in agent infrastructure from Amazon Web Services (AWS), Google, and Vercel let untrusted or forged instructions reach an agent's tools with no check that a model turn had authorized them. In several of the attack paths, the model never ran at all, so system prompts, content filters, and model-level guardrails never got a chance to intervene. The affected products include Amazon

Overview On August 2, 2026, N-able published a security advisory for CVE-2026-18577 , an authentication bypass vulnerability affecting N-central that was discovered being exploited in-the-wild after an incomplete fix for an earlier authentication bypass issue, CVE-2026-18556 was disclosed. CVE-2026-18577 allows a remote unauthenticated attacker to bypass authentication and obtain administrative control of vulnerable N-central servers in affected deployments. N-able N-central is a widely deployed Remote Monitoring and Management (RMM) platform used by managed service providers (MSPs) and…

On August 4, 2026, Elastic Security Labs identified a new Shai-Hulud campaign targeting the maintainer of keyv , a widely used key-value storage library. The attackers trojanized the monorepo and embedded a self-propagating worm called CHAINDROP that uses stolen npm credentials to automatically backdoor every other package the maintainer had publish rights to. The reach of this compromise is significant: keyv alone received over 600 million downloads last month, with related packages compounding the exposure: flat-cache near 580 million, cacheable-request at over 137 million, cacheable at…